Zhibin Lan

zhibinlan's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 41 minutes ago

Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling

Paper • 2512.23959 • Published 5 days ago • 75

New activity in zhibinlan/UME-R1-7B about 2 months ago

CoT data construction

1

#1 opened about 2 months ago by

anaconda123

upvoted a collection about 2 months ago

UME-R1

Collection

UME-R1 is a framework designed to endow multimodal embedding models with the flexibility to switch between discriminative and generative embeddings • 4 items • Updated Nov 4, 2025 • 8

updated 2 models about 2 months ago

zhibinlan/UME-R1-7B

Image-Text-to-Text • 8B • Updated Nov 10, 2025 • 45 • 5

zhibinlan/UME-R1-2B

Image-Text-to-Text • 2B • Updated Nov 10, 2025 • 2.21k • 5

updated 2 datasets about 2 months ago

zhibinlan/UME-sft-train

Preview • Updated Nov 10, 2025 • 42

zhibinlan/UME-rl-train

Viewer • Updated Nov 10, 2025 • 11.5k • 41

liked 2 models 2 months ago

zhibinlan/UME-R1-7B

Image-Text-to-Text • 8B • Updated Nov 10, 2025 • 45 • 5

zhibinlan/UME-R1-2B

Image-Text-to-Text • 2B • Updated Nov 10, 2025 • 2.21k • 5

New activity in TIGER-Lab/MMEB-Leaderboard 2 months ago

Upload 2 files

#76 opened 2 months ago by

zhibinlan

upvoted a paper 2 months ago

UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings

Paper • 2511.00405 • Published Nov 1, 2025 • 5

commented a paper 2 months ago

UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings

Paper • 2511.00405 • Published Nov 1, 2025 • 5 •

1

published 2 models 2 months ago

zhibinlan/UME-R1-2B

Image-Text-to-Text • 2B • Updated Nov 10, 2025 • 2.21k • 5

zhibinlan/UME-R1-7B

Image-Text-to-Text • 8B • Updated Nov 10, 2025 • 45 • 5

published 2 datasets 2 months ago

zhibinlan/UME-rl-train

Viewer • Updated Nov 10, 2025 • 11.5k • 41

zhibinlan/UME-sft-train

Preview • Updated Nov 10, 2025 • 42

updated a collection 2 months ago

UME-R1

Collection

UME-R1 is a framework designed to endow multimodal embedding models with the flexibility to switch between discriminative and generative embeddings • 4 items • Updated Nov 4, 2025 • 8